A subset of cells in glioblastoma multiforme (GBM) has been identified that enjoy a unique capacity to regenerate tumors. These brain tumor stem cells (BTSC) can be segregated by the neural stem cell marker, CD133, and are widely believed to be the cells responsible for resistance to conventional therapies. An effective means of specifically eliminating these cells may reduce the need for intensive and non-specific conventional therapy and lower the risk of tumor recurrence. EGFRvIII is a tumor-specific mutation found on BTSC. We have successfully targeted EGFRvIII using a peptide vaccine that allowed rapid translation to an ongoing Phase III trial. EGFRvIII expression is heterogeneous, however, and the recurrence of EGFRvIII-negative tumors suggests that BTSC can rely on other oncogenic pathways. While our data suggests that targeting tumor-specific mutations in BTSC may be important, few highly-conserved tumor-specific mutations like EGFRvIII will be identified and antigen defined vaccine approaches will ultimately be limited. Dendritic cells (DCs) loaded with amplified total tumor RNA is an innovative strategy to induce cellular and humoral antitumor immune responses. Although CD133(+) BTSC are a minority subpopulation of GBM that cannot be reliably isolated or propagated in sufficient quantities to serve as an antigen source for human vaccination protocols, we have been able to reproducibly amplify the RNA content from as few as 500 sorted CD133(+) tumor cells to generate RNA libraries sufficient for clinical scale DC-based vaccination. In order to focus the immunologic response on antigens preferentially or uniquely expressed within BTSC and limit the potential for autoimmune reactivity against shared antigens expressed in normal cells, we will evaluate approaches to enrich for antigens preferentially or uniquely expressed in BTSC by using full length cDNA affinity based substractive hybridization or an innovative strategy that leverages the ability of the DNA mismatch binding protein, MutS, to isolate cDNAs that contain tumor-specific mutations. These various preparations will be evaluated for differential toxicity and efficacy in an inbred transgenic murine malignant astrocytoma model, in which a subpopulation of CD133(+) tumor cells with BTSC qualities have been identified and CD8(+) and CD4(+) epitopes have been found. If efficacy is seen, the least toxic strategy will be translated into a Phase I study within the context of our existing clinical trial platform. PUBLIC HEALTH RELEVANCE: Treatment for malignant primary brain tumors, which are the most common cause of death among children and account for more deaths in adults than melanoma, currently represents the most expensive medical therapy per quality-adjusted life-year saved currently provided in the United States. A subset of malignant primary brain tumor cells (BTSCs), called brain tumor stem cells, enjoy a unique capacity to regenerate tumors and to resist conventional therapies. In this proposal we will see if targeting antigens preferentially or uniquely expressed by BTSCs will enhance the efficacy and reduce toxicity of immunotherapy.